An intermediate value in the floor 1 linear interpolation was
overflowing
resulting in obvious artifacts on some files.
e.g.
http://upload.wikimedia.org/wikipedia/commons/7/79/Big_Buck_Bunny_small.ogv
Prior to this fix 87 out of 128 64kbit/s mono files decoded with ffmpeg
have
lower PEAQ ODG values than the same files decoded with libvorbis. With
this
fix none of that set have significantly worse ODG values than libvorbis.
Fixes issue 2352
Patch by Gregory Maxwell <greg@xiph.org>
Originally committed as revision 25724 to svn://svn.ffmpeg.org/ffmpeg/trunk
uint16_fast_t is unsigned int (or long) on Linux, which when compared
with int results in an unsigned compare.
Originally committed as revision 24994 to svn://svn.ffmpeg.org/ffmpeg/trunk
When sizeof(uint_fast8_t) >= sizeof(int) there are unintended size effects.
Originally committed as revision 24716 to svn://svn.ffmpeg.org/ffmpeg/trunk
If sizeof uint_fast8_t > 1 and sizeof size_t <= 4, the expression that mallocs
classifs is susceptible to integer overflow.
Originally committed as revision 24675 to svn://svn.ffmpeg.org/ffmpeg/trunk
Passing an explicit filename to this command is only necessary if the
documentation in the @file block refers to a file different from the
one the block resides in.
Originally committed as revision 22921 to svn://svn.ffmpeg.org/ffmpeg/trunk
10_vorbis_submap_indexes.patch by chrome.
I am applying this even though Reimar had some comments to improve it as it fixes
a serious security issue and I do not want to leave such things unfixed.
Originally committed as revision 20001 to svn://svn.ffmpeg.org/ffmpeg/trunk
23_vorbis_sane_partition.patch by chrome.
Also this should be better documented but i prefer not to leave potential
security issues open due to missing documentation.
Originally committed as revision 19996 to svn://svn.ffmpeg.org/ffmpeg/trunk
Now the sign of audio samples in ffvorbis output is the same as in original
uncompressed audio file and this also allows the use of tiny_psnr to compare
ffvorbis with libvorbis/tremor.
Originally committed as revision 18878 to svn://svn.ffmpeg.org/ffmpeg/trunk
Scaling (i)MDCT output has no runtime overhead and can be used to improve
performance of audio codecs. All the changes are only needed in
'ff_mdct_init' function and slow down initialization a bit.
Originally committed as revision 18855 to svn://svn.ffmpeg.org/ffmpeg/trunk
This was off-by-one bug which affects floor0 encoded files. Tracked down
by valgrind (was reported as the use of uninitialized memory).
With this patch applied, tiny_psnr now reports the following when ffvorbis
is compared to libvorbis-1.0beta4 output for 'lsp-test.ogg' sample:
stddev: 0.76 PSNR: 98.69 bytes: 3498380/ 3498452
Output from libvorbis-1.2 is substantially different:
stddev: 132.57 PSNR: 53.87 bytes: 3498424/ 3498496
Originally committed as revision 18821 to svn://svn.ffmpeg.org/ffmpeg/trunk
AVPacket argument rather than a const uint8_t *buf + int buf_size. This allows
passing of packet-specific flags from demuxer to decoder, such as the keyframe
flag, which appears necessary to playback corePNG P-frames.
Patch by Thilo Borgmann thilo.borgmann googlemail com, see also the thread
"Google Summer of Code participation" on the mailinglist.
Originally committed as revision 18351 to svn://svn.ffmpeg.org/ffmpeg/trunk
Otherwise doxygen complains about ambiguous filenames when files exist
under the same name in different subdirectories.
Originally committed as revision 16912 to svn://svn.ffmpeg.org/ffmpeg/trunk
